APPLICANT'S ABSTRACT: Acupuncture substance abuse treatment is being used in many treatment programs and has strong support from several investigators. Our position is that, while the results of the presently available studies are promising, additional well designed studies are needed to help confirm the impression of effectiveness before the method becomes firmly entrenched in substance abuse treatment programming. This proposal will address three basic issues in alcohol acupuncture treatment. 1. What are the relative treatment outcome size effects associated with conventional psycho-social treatment, and true, sham, and electroacupuncture?2. To what extent does placebo response contribute to conventional and acupuncture treatment outcome? 3. Are there significant, gender-related treatment outcome effects? The acupuncture treatment protocol under investigation was established in prior alcoholism research and has been pretested in three preliminary cocaine treatment studies which examined the feasibility of the treatment and placebo methods. A fully randomized, placebo controlled study will be conducted in an established residential and outpatient treatment program. Patients will be randomly assigned to one of four treatment modalities. The design will permit the comparison of true, sham, and electroacupuncture, and conventional treatment alone. All patients will receive conventional treatment. Prior to treatment, all patients will be tested for placebo responsiveness using a procedure developed for pain acupuncture research. Treatment outcome will be measured at the end of each week of treatment, at the end of treatment, and at three post treatment intervals, three months, six months and twelve months. Treatment outcome measures include self report level of craving, the Timeline Follow-Back, the composite scales of the Addiction Severity Index, confirmation using informants and document review, treatment dropout, the Medical Outcome Study Measure, Beck Depression Inventory, and Self-rating Anxiety Scale. Post-treatment utilization of medical, substance abuse and correctional system services will be noted. Measurement raters will be blinded to type of acupuncture treatment. Multivariate analyses will examine the relationship of placebo and nonspecific treatment factors to outcome, treatment effect sizes, relapse rates, and predictors of treatment outcome.